Watts pointed out that the weight he had invested in guns could be used differently. Putting the weight added for four 9.2in into armour would buy thicker main side armour – 10in and 8in and a 7in battery instead of 9in and 7in and 4in. Alternatively, it could provide more power (19,000 IHP) and higher speed (19 knots).
Lord Nelson shows the 3pdrs atop her after 12in and 9.2in turrets, with 12pdrs in her superstructure.
In Design C the heavy armament (four 12in, eight 9.2in) was concentrated in triangular ‘bunches’ fore and aft, maximising end-on fire. Concentration also made it possible to reduce armour weight by enclosing the bases of three turrets in a single armoured redoubt. The usual 6in battery was eliminated. Watts estimated that a Design C ship would cost £1,190,000. Design D was C with an additional 9.2in turret on each broadside, for a total of twelve such guns (he claimed that six of them could fire end-on). An additional novel feature of Designs C and D was that the 12in guns fired over the 9.2in gunhouses. All of these ships had the same anti-torpedo (boat) battery as King Edward VII: fourteen 12pdrs backed by ten 3pdrs.
In October and November Watts offered enlarged versions of B with more armour and thicker shields, with greater freeboard etc. B1 was a dramatic step towards an all-big-gun ship, the 6in battery being abolished (B2, however, had four 7.5in guns in addition to the 9.2in, in gunhouses).13 A Legend dated October 1902 offered alternatives B3 and B4 with, respectively, eight and ten 9.2in guns. It seems likely that the 6in battery was eliminated to bring the ship’s size down to something considered affordable. Trial speed was set at 18 knots. Displacement was 16,000 and 16,100 tons. Maximum belt thickness was 12in.
The 19 November offering (Design E) was an all-big-gun ship, perhaps the first the Royal Navy ever considered: twelve 10in guns, with nothing else but anti-torpedo weapons (fourteen 12pdrs and ten 3pdrs), plus the usual torpedo tubes (four on the broadside, one at the stern).14 Armament arrangement was fixed by the ship’s reciprocating powerplant: one turret at each end, two on each side, for an eight-gun broadside. Probably not coincidentally, E was roughly what Admiral Fisher was soon proposing. The 10in gun was specified because it seemed to be the largest which could be fired rapidly. Compared to the October designs, this one would have displaced somewhat less (15,800 tons). Speed would have been the same (18 knots on trial, 16.5 knots continuously, on 15,500 IHP). The origin of this all-big-gun design is not evident from the Cover.
Watts offered modified versions of the B designs, with eight, ten or twelve 9.2in guns – and no 6in battery – in December 1902. He submitted a Legend for a ‘New Design’ (16,000 tons, 18 knots) armed with four 12in and ten 9.2in on 11 February 1903. It would have had two twin and one single 9.2in mountings on each broadside and designs for both twin and single mounts were requested from Vickers and Elswick. According to a note in the Cover (24 February 1903, by DNC’s deputy Mr Narbeth), the 16,000 ton B3d design (ten 9.2in) was approved with changes: (i) an 8in citadel with 2in splinter protection, as on the main deck; (ii) twelve rather than ten 9.2in guns, the extra weight to come out of the margin and total ammunition supply not being increased (it would be 1000 rounds of 9.2in, 100 each for ten guns); (iii) machinery spaces would be unchanged; (iv) the 9.2in guns would be arranged to fire further across the beam (by raising them); (v) the ship might be lengthened to bear the extra weight due to lifting the forward 9.2in guns and to maintain the ship’s speed without extra power.15 By June, detail design was proceeding. For example, pending the report of the Boiler Committee, it was decided to give the ship one-fifth cylindrical boilers and four-fifths small-tube (Babcock & Wilcox) boilers. Consideration was given to using new 12in/45s in place of the existing 12in/40s (adding about 100 tons). The ships were also given the new 9.2in/50s, which were heavier than the earlier type. There was also interest in providing them with full ammunition. Watts’ suggestion that weight be saved by adopting fast-running engines was rejected.
In May, Controller ordered that the design be completed in time for contracts to be placed in October 1903 under the 1903–4 programme.16 The Cover includes an approved Legend dated 4 August 1903 (originally dated 30 July). The design received the Board Stamp on 6 August. DNC must have been shocked when the Board suddenly withdrew approval of the design a few days later.17 Without any approved new design, three repeat King Edward VIIs had to be built under the 1903–4 programme.
Controller offered Watts several reasons for the rejection. It was too large (150 tons larger than the King Edward VII class). DNC pointed out that in each case the object had been to carry the most powerful armament possible into action with suitable protection and manoeuvring power on about the same dimensions. Not counting armament, the 1903 battleship would weigh 13,135 tons, compared to 13,775 tons for King Edward VII without its armament. On 640 tons less (without armament), Watts had managed to carry 790 tons more armament as well as far better protection. He argued that his August 1903 design was at least 50 per cent cheaper than a King Edward VII, counting the total of armament and protection, as the latter might be traded for even more armament. First Sea Lord also told DNC’s assistant, Mr Deadman, that approval for the August 1903 design had been cancelled because of limited docking facilities (this was later retracted). Watts argued that the new design had the same clearances of basins and channels as that which had been accepted for the King Edward VII class.
Agamemnon in 1918, little changed except for having had her funnels raised. She was converted after the war into a radio-controlled target; after being discarded in 1926 she was replaced by HMS Centurion.
Watts cited ‘a rumour in the Office’ that the design had been cancelled because the advance in firepower was too great: ‘it is said that it is not considered politic to make so large an advance on what other Powers are doing. I do not believe . . . this.’ Controller had mentioned a further consideration which he was not at liberty to disclose without the Board’s permission; but without any explanation, Watts could not hope to develop acceptable designs. Word that the design had been abandoned leaked outside the Admiralty.18
The solution was apparently to retreat to ten 9.2in guns, the middle mountings on each side being single. A new Legend was dated 19 October 1903. The Cover includes a developed version designated G5, whose Legend is dated 13 November 1903. The design was still in flux. In December 1903 Controller assembled a précis of what had been done in the design of the 1904–5 ships for First Lord Selborne. Meanwhile he asked DNC to look into the possibility of thickening the armour deck at the expense of the upper deck; he also wanted to know how much weight would be saved by going to all water-tube boilers. DNC offered to thicken the flat part of the armour deck to 1¾in by omitting armour gratings and limiting the upper deck between the machinery casings to ½in (and elsewhere to ¾in instead of 1in). He might also be able to gain weight by reducing turret roof thicknesses to those in the King Edward VII class. The Naval Lords decided to adopt all water-tube boilers and also to reduce some armour thickness so as to extend side armour over the armoured deck fore and aft.
The abortive design armed with twelve 9.2in guns showed that there was sufficient space for sixteen heavy guns. The alternatives of sixteen 10in or twelve 12in appear as early as 6 August 1903 in a Legend titled ‘New Battleship Designs: Alternative Armament’. A ship mounting the 12in guns would be perhaps a little longer than one armed with sixteen 10in, but not much. DNC produced a Legend for a ship